PayPal Yang Zhou recalls the 2022 storm: What caused the turmoil that swept everything from Luna to FTX?
The peak of PayPal was from March to April 2021, coinciding with the first surge of the last Bitcoin bull market. That market peak was mainly driven by the Coinbase listing, with Bitcoin reaching a high of $64,000 in April 2021.Author: Wu Says Blockchain
Babel Finance is a leader in the cryptocurrency lending and wealth management sector. During the 312 period in 2020, Babel Finance was on the brink of bankruptcy due to its high-risk operating model, but was saved by Tether's grace and a market rebound. On May 26, 2022, it announced the completion of a Series B financing round, raising $80 million and achieving a unicorn status with a valuation of $2 billion. On June 17, 2022, Babel informed its partners: we are insolvent. The official website stated that it had suspended product redemptions and withdrawal functions. Yang Zhou is a co-founder of Babel Finance but had largely exited by 2021; the video interviewer is Bill, authorized by Wu Says for reprint.
Yang Zhou: Babel's peak was from March to April 2021, coinciding with the first surge of Bitcoin's last bull market. That market peak was mainly driven by Coinbase's listing, with Bitcoin reaching a high of $64,000 in April 2021. At the same time, the funding fee arbitrage in the market also peaked. During that phase, the total assets managed by Babel (including lending and managed funds) were about $7 billion.
The scale of centralized lending institutions has always surpassed that of decentralized DeFi lending protocols. We do not consider Tether (USDT) because its data is opaque. However, according to public information, the largest lending institution at that time was Genesis, with an asset management scale (AUM) of $13.8 billion, far exceeding the current AAVE and other DeFi lending protocols.
At that time, the main customer groups of centralized lending institutions included:
● Asian crypto miners (holding large amounts of BTC assets, needing liquidity);
● U.S. institutional investors (mainly engaging in GBTC arbitrage trading).
Asian miners benefited from the rise of Bitcoin, with their assets growing 20-30 times in just a few years. Initially, their assets might have only been $100-200 million, but during the bull market, they quickly appreciated to $4-5 billion.
The growth momentum in the U.S. market mainly came from GBTC (Grayscale Bitcoin Trust) arbitrage trading. From 2018 to early 2021, GBTC maintained a premium, forming a stable arbitrage model in the market:
Investors borrow BTC to subscribe to GBTC, locking it for 6 months;
At the same time, they sell BTC in the spot market for hedging;
After 6 months, they unlock GBTC and sell it for profit.
Bill: Yes, the 6-month arbitrage window for GBTC was one of the main lending scenarios for Genesis at that time.
Yang Zhou: Yes, Genesis needed BTC supply, while institutions in the Asian market needed U.S. dollars. Therefore, Genesis provided us with U.S. dollar loans, and we provided BTC to them, forming a stable arbitrage loop, satisfying each other's liquidity needs. This model naturally scaled up in a market environment where BTC prices were rapidly increasing.
However, this growth was mainly driven by market Beta (i.e., overall industry growth), rather than pure institutional expansion. The core factor was the rise in BTC prices themselves, as the total supply of BTC did not change dramatically, but demand surged in the short term.
Bill: After understanding Babel's early development, I also know that you later started proprietary trading, which was also the starting point for the subsequent risk explosion. Can you share the background here?
Yang Zhou:
Babel's development can be divided into three stages:
- Pure lending business phase (2018-2020)
● Miners used BTC as collateral to borrow stablecoins (USDT, USDC);
● The business model was simple, mainly serving miners and institutions.
- Entering wealth management and asset management business (2020-2021)
● Due to market demand, Babel began offering BTC or ETH-based wealth management products;
● Using options structures to achieve returns, such as selling CALL/PUT structured products.
- Proprietary trading and leverage expansion phase (2021-2022)
● In an imperfect regulatory environment, lending, asset management, and proprietary trading were mixed together;
● Due to the rise in BTC prices, market liquidity was abundant, and trading strategies were aggressive.
Wall Street financial institutions have actually gone through a similar process:
● Initially, lending, asset management, and proprietary trading were mixed;
● After regulatory intervention, they were split, but after a while, they began to merge again;
● Repeatedly experiencing cycles of merging-splitting-risk explosion.
For the crypto industry, this cycle is shorter, with a major crisis occurring almost every 3-4 years, while traditional finance may have a cycle of 10-20 years.
Bill: Which position ultimately led to the risk that caused the liquidity crisis for Babel?
Yang Zhou: To truly trace the source of the risk, we have to go back to the high funding fee market at the beginning of 2021.
● At that time, the funding fee arbitrage in the market maintained an annualized rate of 40-50%, attracting a large influx of funds;
● The supply of USDT surged from $20 billion (end of 2020) to $60 billion (May 2021);
● Due to the influx of funds into arbitrage trading, many institutions launched high-yield products with 1-2 year terms, leading to funds being locked up.
In the fourth quarter of 2021, the market entered a bear market (due to the Federal Reserve's tightening monetary policy), resulting in:
● A decline in funding demand, but funds could not be quickly withdrawn;
● Funds began to seek new yield scenarios, such as the Anchor protocol (UST & Terra).
Ultimately, the Terra/Luna system collapsed in May 2022, causing:
● $40-50 billion of base currency to disappear;
● The cryptocurrency market's total market value evaporated by $600 billion;
● Alameda (the market maker under FTX) took on the market liquidity crisis.
Bill: So, when Alameda took over Terra/Luna, it became the largest loss bearer in the market, right?
Yang Zhou: Yes, Alameda's losses could be between $10-20 billion, but because it is a centralized institution, it could hide losses through some means, so the outside world did not notice its problems in a short time. However, the chain reaction caused by the collapse of Terra/Luna transmitted to Three Arrows Capital (3AC), Babel Finance, BlockFi, and other institutions.
Bill: How did this transmission effect specifically occur?
Yang Zhou: At that time, there were not many funding exits for centralized lending institutions, and there were only a few large institutions truly providing lending in the industry. Institutions that dared to borrow money at an annual interest rate of 7-8% had even more singular funding flows.
By late 2021, we found that a large amount of lending funds flowed directly or indirectly into the Terra ecosystem.
Bill: So, you were actually putting borrowed money into the Terra protocol for wealth management?
Yang Zhou: Yes, directly or indirectly, these funds were ultimately linked to the Terra ecosystem. When Terra collapsed, all centralized lending institutions in the cryptocurrency industry faced enormous risks.
Bill: Wait a minute, in other words, in the second half of the 2021 bull market, the flow of lending funds was no longer for miners to produce or mine, but rather for more people to borrow money for speculation, ultimately leading to a large influx of funds into Luna's UST peg mechanism, right?
Yang Zhou: Correct. In fact, miners began to reduce leverage after Bitcoin rose to $20,000 in December 2020. By the time Bitcoin prices reached $30,000 and $40,000, they had already reduced borrowing and started selling coins to realize profits.
Bill: Miners are actually more risk-aware, right?
Yang Zhou: Exactly. Miners have experienced too many bull and bear cycles. For example, the reason Bitcoin prices have struggled to break through the $100,000 mark is partly because miners have been continuously selling around that level. Many miners I know reduced their positions around $100,000.
In the previous cycle (2021), the crash around $40,000 was also similar to the situation around $100,000 in this cycle—both were caused by miners selling large amounts of coins, putting pressure on the market.
By the end of 2021, the flow of lending funds had completely shifted, mainly used for funding fee arbitrage rather than miners' production needs. Our lending portfolio had over half of its funds entering financing arbitrage trading, and these arbitrage trades largely depended on UST (Terra ecosystem) stablecoin yields.
Bill: After the collapse of Terra, your borrowers' investment funds were also completely gone. How did the subsequent chain reaction occur?
Yang Zhou: At that time, the overall market risk had not fully manifested. The first to be affected were a few institutions with the highest leverage, whose positions blew up first.
● Three Arrows Capital (3AC)
● Babel Finance
● BlockFi
From a market trading perspective, the first to fall were 3AC and Babel, as these two institutions were the largest options market makers and short volatility traders in the market.
Their strategy was to simultaneously sell PUT and CALL options, but as the market plummeted, the losses from selling PUTs far exceeded the profits from selling CALLs.
Bill: So these institutions were not really trying to bottom fish when the market fell, but were instead engaging in arbitrage? But when the market collapsed in a short time, they couldn't bottom fish at all?
Yang Zhou: Yes. If the market slowly dropped by 40% (for example, over 2-3 weeks), then options traders could still hedge risks or supplement positions through margin calls.
However, the market plummeted by 40% in just two or three days, leaving these institutions unable to hedge in time.
At that time, the situation was:
● The dynamic hedging mechanism in the options market had failed;
● BTC prices fell from $50,000 to $35,000, then quickly to $20,000;
● A large number of options sellers were forced to liquidate, increasing market selling pressure and creating a death spiral.
Bill: So this is similar to the Gamma Squeeze mechanism seen in recent Bitcoin price increases? Only this time it was the reverse— the faster the market fell, the more institutions had to sell, further exacerbating the market collapse?
Yang Zhou: Exactly. When the market plummeted, all market makers and options traders had to sell BTC to hedge, ultimately causing the market to fall even faster, creating a vicious cycle.
This led to:
● The bankruptcy of Three Arrows Capital
● The bankruptcy of Babel Finance
● The bankruptcy of BlockFi
Bill: Did Three Arrows Capital and Babel Finance go bankrupt at the same time?
Yang Zhou: Essentially at the same time, but 3AC acted first. Their founder, Su Zhu, as a professional trader, was quicker to cut losses than Babel.
However, the bankruptcies of 3AC and Babel affected more counterparties, leading to:
● Other institutions beginning margin calls, further causing panic.
● Other lending institutions like BlockFi and Voyager also exposed problems.
Bill: This was already June-July 2022, right?
Yang Zhou: Yes, in June-July, this crisis began to spread to larger institutions, and Alameda was already deeply mired in it.
At that time, Alameda was borrowing heavily from BlockFi and Voyager, so FTX chose to step in to rescue these institutions, attempting to cover up Alameda's losses.
However, everything collapsed completely after CZ tweeted questioning FTX.
Bill: So, if SBF had been able to pause withdrawals at that time, would it have been possible to hold on?
Yang Zhou: Perhaps. But history has no ifs. If FTX had chosen to pause withdrawals like OKX, it might have been able to hold on. But SBF chose to confront the situation head-on, and ultimately FTX collapsed directly.
If the U.S. government had been willing to step in and rescue the market like traditional finance, perhaps FTX would not have collapsed so quickly. But the reality is that SBF was merely a cash cow for politicians, not a truly important financial institution in their eyes.
Yang Zhou: After FTX collapsed, Genesis also couldn't hold on. Genesis was the largest lending counterparty in the industry (AUM $13.8 billion).
Bill: With Genesis's asset management scale (AUM) reaching $13.8 billion, where did its funding mainly come from? Was it also from Tether?
Yang Zhou: No. Genesis's funding sources are very diversified, with a large number of credit fund clients, including many traditional financial institutions.
Bill: So, its funding sources are relatively diversified?
Yang Zhou: Yes, and many of Genesis's funding sources are traditional institutions outside the crypto industry, which is one of its unique aspects. This also means that the problems Genesis encountered ultimately transmitted to traditional financial institutions, indirectly leading to the bankruptcy of several crypto banks in 2023.
However, the bankruptcy of these banks was not directly triggered by risks in the crypto market, but rather because the Democratic government at the time directly ordered the closure of some crypto banks. This was not because these banks were already facing risks, but more like a forced administrative action.
Yang Zhou: In my view, the main motivation for the Democratic Party to take this measure at that time was to quickly distance themselves from FTX. Because FTX had made a lot of money in the past few years and had close relationships with some political figures, when FTX exploded, they might have been eager to cut ties to avoid deeper entanglements.
At the same time, the crisis at Silicon Valley Bank (SVB) was actually unrelated to cryptocurrency, but rather a general interest rate risk facing the entire banking system.
Bill: SVB's bankruptcy was due to problems on its asset side, right? It held a large amount of U.S. Treasury bonds, and rising interest rates caused the market value of these bonds to plummet.
Yang Zhou: Exactly. At that time, SVB and other affected banks held a large number of long-term bonds, but these bonds lost market value as interest rates rose, resulting in significant paper losses. Although these bonds would not have been a problem if held to maturity, due to large withdrawals by customers, the bank had to sell bonds to obtain liquidity, leading to actual losses.
The reason for USDC's depegging was that 10% of USDC deposits were held at SVB. After SVB's bankruptcy, the market began to panic, fearing that USDC's asset side might not fully cover its issuance, leading to a brief depegging, with USDC's price dropping below $0.90.
Bill: So, SVB's bankruptcy was not because it held crypto assets, but due to a debt crisis in the traditional financial system?
Yang Zhou: Yes, this is largely unrelated to the crypto market. But this incident also exposed the fact that the stablecoin issuing institutions in cryptocurrency had not fully integrated into the banking risk management system. At that time, many stablecoin institutions had not fully realized that banks themselves also faced credit risks.
Bill: But in the end, the Federal Reserve (FED) intervened and printed $500 billion to stabilize the banking system, right?
Yang Zhou: Yes, the Federal Reserve ultimately stepped in to rescue the market, injecting $500 billion into the banking system. In comparison, the overall scale of the crypto market is even smaller than this rescue fund. But this also exposes a problem:
● When traditional financial institutions encounter problems, the Federal Reserve will rescue them;
● When cryptocurrency institutions encounter problems, they can only fend for themselves.
If Tether had stepped in to rescue FTX at that time, perhaps FTX would not have collapsed. But Tether is essentially a private institution; it is not a public institution like the Federal Reserve, so it does not have the motivation to rescue FTX.
Bill: Financial crime is a felony in the U.S., but the largest financial crime in history was actually the 2008 CDS (credit default swap) crisis, right?
Yang Zhou: Yes, the core issue of the 2008 financial crisis was that a large number of subprime loans were packaged by financial institutions into high-rated assets and then traded with multiple layers of leverage, ultimately forming systemic financial fraud. But because this crisis was too large and affected the global economy, the Federal Reserve and the government ultimately chose to rescue the market, while the real perpetrators were hardly punished.
You can look at the movie "The Big Short"; its final conclusion is this reasoning. Ultimately, the political system in the U.S. and globally still faces the same problem:
● When the crisis is big enough, the government will step in to rescue;
● If you are just a small pond, then sorry, you can only fend for yourself.
Bill: So, we can see that the U.S. government, when facing political enemies, is also starting to learn China's "anti-corruption" strategy?
Yang Zhou: Yes, the U.S. government is now also starting to use "anti-corruption" as a means of political struggle. Just as China learns from the West in developing AI, the U.S. is learning from China’s strategies in certain aspects. But for the crypto industry, the scale of the last cycle was still not large enough, so the government was unwilling to rescue it.
Bill: Looking back at the last cycle, from Babel Finance to FTX, do you think this cycle will have similar systemic risks?
Yang Zhou: I believe the situation in this cycle will be much better.
After the last crisis, many crypto institutions significantly strengthened their risk controls, mainly reflected in several aspects:
- More institutions began to use custody services
● For example, Binance now has Ceffu custody services, which, although related to Binance, at least manages assets off the balance sheet.
● Exchanges like OKX and Coinbase have also strengthened their custody services.
- The business of exchanges and lending institutions is gradually separating
● Traditional financial institutions took decades to undergo robust transformations, while the crypto industry completed similar adjustments in just two years.
- Upgraded awareness among market participants
● After several rounds of crises, the entire crypto industry has evolved faster, and institutions and investors have a stronger risk awareness than before.
● This makes me feel that the crypto industry has a particularly strong adaptability; despite being full of speculation, it continues to evolve and move forward.
Bill: Indeed, the entire crypto industry has evolved faster than traditional finance.
Bill: When Babel exploded, you had already retired and left the company. Why did you later choose to return and try to save Babel?
Yang Zhou: Many people have asked this question. Some might think I had some "bargaining chip" with Babel, so I had to come back. But the truth is, I had no issues at all; it was purely my personal character.
At that time, I believed the most correct choice was to try to save the company. I wanted to see if it was possible to restructure the debt to salvage the situation.
Babel still had:
● 15,000+ BTC at Genesis;
● 20,000-30,000 ETH;
● $200 million in cash.
In that situation, I felt it was possible to save it, but ultimately Genesis also collapsed, leading to these assets being unrecoverable.
Bill: So, you essentially wrote a Call Option for all creditors?
Yang Zhou: Yes, a Call Option on the debt.
